Hi. I just read through the original study from which you derived this graph. I think you may have misinterpreted the data in a big way. The original graph in Gruèrea & Sengupta (2011) is Figure 1, which shows the cumulative number of suicides (~100k for all of India, ~20k for farmers), not the rate of suicide per 100,000 people-year. It appears that you simply divided both the 100k and 20k numbers by the population of india and multiplied by 100k to get the rates in your graph. This is not a correct method. When finding the rate of death in a subpopulation, such as for farmers, you would divide by the total number of farmers. For the year 2011 this was estimated (by Times of India) to be 262 million cultivators (118 million) and ag laborers (144 million). Gruèrea & Sengupta (2011) do not distinguish between cultivators and ag labor, and do not give a total occupation population for the subgroup that they are studying -- which is rather careless of them to leave that out. Regardless, to make your graph, you should have divided the ~20k farmer suicide number by either 262 million or 118 million and then multiplied by 100k. That would yield, respectively, a rate of ~8 suicides per 100k cultivators and ag laborers, or ~17 suicides per 100k cultivators only. As you can see, there are multiple sources of error and ambiguity at play here, but we can be sure that farmer suicides in India are at least 8 if not 17 per 100k, and not below 2 per 100k, as your chart suggests.
Tarnas ( talk) 00:01, 7 September 2014 (UTC)
Hi. I just read through the original study from which you derived this graph. I think you may have misinterpreted the data in a big way. The original graph in Gruèrea & Sengupta (2011) is Figure 1, which shows the cumulative number of suicides (~100k for all of India, ~20k for farmers), not the rate of suicide per 100,000 people-year. It appears that you simply divided both the 100k and 20k numbers by the population of india and multiplied by 100k to get the rates in your graph. This is not a correct method. When finding the rate of death in a subpopulation, such as for farmers, you would divide by the total number of farmers. For the year 2011 this was estimated (by Times of India) to be 262 million cultivators (118 million) and ag laborers (144 million). Gruèrea & Sengupta (2011) do not distinguish between cultivators and ag labor, and do not give a total occupation population for the subgroup that they are studying -- which is rather careless of them to leave that out. Regardless, to make your graph, you should have divided the ~20k farmer suicide number by either 262 million or 118 million and then multiplied by 100k. That would yield, respectively, a rate of ~8 suicides per 100k cultivators and ag laborers, or ~17 suicides per 100k cultivators only. As you can see, there are multiple sources of error and ambiguity at play here, but we can be sure that farmer suicides in India are at least 8 if not 17 per 100k, and not below 2 per 100k, as your chart suggests.
Tarnas ( talk) 00:01, 7 September 2014 (UTC)